CN106097256B - A kind of video image fuzziness detection method based on Image Blind deblurring - Google Patents
A kind of video image fuzziness detection method based on Image Blind deblurring Download PDFInfo
- Publication number
- CN106097256B CN106097256B CN201610373160.XA CN201610373160A CN106097256B CN 106097256 B CN106097256 B CN 106097256B CN 201610373160 A CN201610373160 A CN 201610373160A CN 106097256 B CN106097256 B CN 106097256B
- Authority
- CN
- China
- Prior art keywords
- image
- blind
- deblurring
- fuzzy
- video image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 28
- 238000000034 method Methods 0.000 claims abstract description 19
- 230000011218 segmentation Effects 0.000 claims abstract description 10
- 238000009795 derivation Methods 0.000 claims description 3
- 238000003384 imaging method Methods 0.000 claims description 3
- 238000012986 modification Methods 0.000 claims description 3
- 230000004048 modification Effects 0.000 claims description 3
- 238000000638 solvent extraction Methods 0.000 claims description 2
- 238000002372 labelling Methods 0.000 abstract 1
- 230000035939 shock Effects 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 3
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000001303 quality assessment method Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/73—Deblurring; Sharpening
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20192—Edge enhancement; Edge preservation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30168—Image quality inspection
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The video image fuzziness detection method based on Image Blind deblurring that the invention discloses a kind of, it is specific as follows: (1) firstly, a frame fuzzy video image is passed through blind 1 method of deblurring, to obtain clear image f1.(2) then by clear image f1It carries out conspicuousness detection and obtains notable figure.(3) notable figure is subjected to connected area segmentation again, then carries out connected component labeling, mark region corresponding with the connected domain of clear image in input blurred picture.(4) the structural similarity value that blurred picture and each corresponding connected region of clear image are calculated using structural similarity index is calculated weighted average and obtains S1.(5) input blurred picture is passed through into blind 2 method of deblurring, obtains clear image f2.(6) step (2) are repeated and arrives (4), obtain S2.(7) to S1And S2Different weights is assigned, the fuzziness of input picture: M is finally obtainedblur=ω1S1+ω2S2.The present invention keeps fuzziness detection more accurate.
Description
Technical field
The present invention relates to digital image processing techniques field, especially a kind of video image mould based on Image Blind deblurring
Paste degree detection method.
Background technique
Fuzziness is an important measurement index of picture quality, and the ambiguity evaluation of image is image quality evaluation neck
The important subject in domain.Many existing ambiguity evaluation methods are established in the design feature of image to be evaluated itself, than
Such as marginal information, textural characteristics, such methods are confined to the fuzziness of more identical content images, and there are also can be used for
The method for evaluating different content image blur, but complexity is higher, and real-time is poor.Full reference picture ambiguity evaluation method
In be difficult to obtain undistorted image, single blind deblurring algorithm accuracy is not high.
Summary of the invention
One kind is provided the technical problem to be solved by the present invention is to overcome the deficiencies in the prior art, mould is gone based on Image Blind
The video image fuzziness detection method of paste, the present invention combines two kinds of blind deblurring algorithms for being directed to different vague category identifiers, and divides
Different weights are not assigned, finally make fuzziness detection more accurate, it can be preferably adaptive for different type blurred picture
Detect fuzziness.
The present invention uses following technical scheme to solve above-mentioned technical problem:
A kind of video image fuzziness detection method based on Image Blind deblurring proposed according to the present invention, including it is following
Step:
Step 1: one frame fuzzy video image of input, carries out two distinct types of Image Blind deblurring, respectively obtain clear
Clear image f1、f2；
Step 2: by clear image f1、f2Conspicuousness is carried out respectively to detect to obtain notable figure, and notable figure is subjected to connected domain
Segmentation, and in clear image f1、f2In mark each connected region position；
Step 3: by the fuzzy video image of input and clear image f1、f2Middle connected region same position is marked；
Step 4: calculating separately the fuzzy video image and clear image f of input1、f2The structure of corresponding connected region
Similarity SSIMi、SSIMj；Wherein, i is integer and 1≤i≤N, N are that input picture through the Image Blind of the first seed type removes mould
The finally obtained connected region number of formulating method, j is integer and 1≤j≤M, M are Image Blind of the input picture through second of type
The finally obtained connected region number of deblurring method；
Step 5: calculating the SSIM of all areasiWeighted average S1, SSIMjWeighted average S2,And different weights are assigned as the fuzzy of input blurred picture from S2 to S1
Degree.
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the Image Blind deblurring of the first seed type is specific as follows in the step 1:
(1), blind fuzzy core is carried out with image deblurring model and its algorithm based on L1/L2 sparse prior first to estimate
Meter；
(2), it obtains carrying out deblurring using quickly non-blind deblurring algorithm after blind fuzzy core, obtains clear image f1。
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the Image Blind deblurring of second of type is specific as follows in the step 1:
(1), motion blur algorithm ambiguous estimation core is removed with Fast Blind first；
(2), the non-blind deblurring method of image based on super Laplace prior is recycled, clear image f is obtained2。
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the blind fuzzy kernel estimates further comprise:
(1), derivation is carried out to input fuzzy video image x, obtains high frequency imaging y；
(2), the blind deblurring model modification f constrained based on L1/L2 regular expression sparse prior is utilized1:
Wherein, k is blind fuzzy core to be estimated, k=[k1,k2...], kiFor the component of blind fuzzy core k, ki≥0,∑iki=
1, α, β is nonnegative number,It is f1Derivative,It is1 norm and 2 norms ratio；
F is updated using alternating projection iterative algorithm1, that is, optimize:
(3), weight least square method is assigned again using no constraint iteration and updates blind fuzzy core, that is, optimize:
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the fuzzy kernel estimates further comprise:
(1), first with the noise in two-sided filter removal image；
(2), edge enhancing is carried out to image using shock filter again, utilizes useful marginal information ambiguous estimation core.
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the conspicuousness detection method obtain notable figure using GBVS algorithm；The connected area segmentation algorithm is adjacent using 8
Then the binaryzation connected domain figure that segmentation obtains is carried out convolution with obtained clear image and obtains connected region by regional partition algorithm
Domain figure finally marks out regional location corresponding with the connected region figure in input fuzzy video image.
It is further excellent as a kind of video image fuzziness detection method based on Image Blind deblurring of the present invention
Change scheme, the fuzziness M of the input blurred pictureblurAre as follows: Mblur=ω1S1+ω2S2, ω1、ω2For different weights.
The invention adopts the above technical scheme compared with prior art, has following technical effect that the present invention combines two kinds
For the blind deblurring algorithm of different vague category identifiers, and different weights are assigned respectively, finally make fuzziness detection more accurate, it can
To be preferably directed to different type blurred picture self-adapting detecting fuzziness.
Detailed description of the invention
Fig. 1 is inventive algorithm flow chart.
Fig. 2 is structural similarity detection algorithm flow chart.
Fig. 3 is blind 1 algorithm flow chart of deblurring.
Fig. 4 is blind 2 algorithm flow chart of deblurring.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with the accompanying drawings and the specific embodiments
The present invention will be described in detail.
Fig. 1 is inventive algorithm flow chart, a kind of video image fuzziness detection method based on Image Blind deblurring, packet
Include following steps:
Step 1: one frame fuzzy video image of input, carries out two distinct types of Image Blind deblurring, respectively obtain clear
Clear image f1、f2；Wherein, f1For Image Blind deblurring (referred to as blind deblurring 1) the obtained clear figure for carrying out the first seed type
Picture, f2For Image Blind deblurring (referred to as blind deblurring 2) the obtained clear image for carrying out second of type.
The Image Blind deblurring of first seed type further comprises:
(1) blind fuzzy kernel estimates are carried out first.
(2) it obtains carrying out non-blind deblurring using RL Deconvolution Algorithm Based on Frequency after fuzzy core, obtains clear image f1。
Preferably, Fig. 3 is blind 1 algorithm flow chart of deblurring, and the blind fuzzy kernel estimates further comprise:
(1) derivation is carried out to input blurred picture x, obtains high frequency imaging y.
(2) the blind deblurring model modification f based on L1/L2 regular expression sparse prior is utilized1:
Wherein, k is blind fuzzy core to be estimated, k=[k1,k2...], ki≥0,∑i ki=1, α, β are non-negative, are used for canonical
With last.f1For finally obtained clear image.Wherein,It is1 norm and 2 norms ratio
Value, as image is increasingly fuzzyyer, which is gradually increased, and updates f using iterative algorithm by this bound term1.That is: it utilizes
Section 2 in formula (1) constrains first item.Last in formula (1) is used to inhibit noise.
F is updated using alternating projection iterative algorithm1, that is, optimize:
Wherein,It is1 norm and 2 norms ratio, as image is increasingly fuzzyyer, the ratio
It is gradually increased.
(3), weight least square method is assigned again using no constraint iteration and updates blind fuzzy core, that is, optimize:
Referring specifically to " Blind deconvolution using a normalized sparsity measure "
(Krishnan D and Fergus R)In CVPR,2011:233-240.
Preferably, if Fig. 4 is blind 2 algorithm flow chart of deblurring, the Image Blind deblurring 2 further comprises:
(1) first with the noise in two-sided filter removal image.The two-sided filter of use are as follows:
Wherein, IeImage is exported for two-sided filter, q is normalization factor, and e is target pixel points, peFor the picture of pixel e
Element value.θ is the neighborhood centered on e, e0It is neighborhood territory pixel point, the space length between two pixels of Q (*) function representation, G
The weight of similarity degree between two pixel of (*) function representation.When Q (*) is larger, weight G (*) is smaller, in image border part,
Two Difference of Adjacent Pixels are larger, so weight is smaller, then the influence of the pixel for pixel point outside edge is smaller, so, Ke Yi
Retain edge when removing noise.
(2) edge enhancing is carried out to image using shock filter again, utilizes useful marginal information ambiguous estimation core.Its
In, the mathematical model of shock filter is as follows:
Wherein, output is L after shock filter, and X is the horizontal direction in space, and Y is the vertical direction in space, LXFor
The derivative of horizontal direction, LYFor the derivative of vertical direction, t is the time,Indicate that the single order direction of image L is led
Number, Δ L=LX 2LXX+2LXLYLXY+LY 2LYY, it is the Second order directional of image L.Final L, initial value are obtained by continuous iteration
For the output of two-sided filter.Extraction for useful edge, as follows using model:
Wherein, L indicates the output blurred picture of previous step,Indicate the derivative of the pixel at v, Nh(e) indicate with
The window that size centered on pixel e is h*h illustrates it is flat site or spike at v when r (v) is smaller.So in order to
Enhance edge, threshold value can be set and remove lesser r (v).
(3) by shock filter enhance edge and after extracting useful edge, then carry out fuzzy kernel estimates, utilize with
Drag:
Wherein,It is the clear image edge obtained by previous step, k is the fuzzy core finally estimated,It is fuzzy
The gradient namely edge of image x, η is non-negative, and then continuous iteration obtains final fuzzy core.
(4) the non-blind deblurring method of image based on super Laplace prior is finally utilized, clear image f is obtained2.Specifically
Referring to document " Fast image deconvolution using hyper-laplacian priors " (D.Krishnan
jandR.Fergus),Advances in Neural Information Processing Systems,2009,vol.22,
pp.1033-1041.
Two width clear images are carried out conspicuousness to detect to obtain notable figure respectively, notable figure is subjected to connected area segmentation, and
Each connected region position is marked in clear image.
Preferably, the conspicuousness detection method obtains notable figure using GBVS algorithm.The connected area segmentation is calculated
Method uses 8 neighborhood partitioning algorithms.Then, the binaryzation connected domain figure that segmentation obtains convolution is carried out with obtained clear image to obtain
To connected region figure.It is finally similarly inputted in blurred picture in two width respectively and marks region corresponding with the connected region figure
Position.
The connected region opposite position marked in blurred picture and clear image will be inputted to mark.
The structural similarity of input blurred picture connected region corresponding with clear image is calculated, finally, calculating all
Fuzziness of the weighted average of the structural similarity of connected region as input picture.
Fig. 2 is structural similarity detection algorithm flow chart, and the calculating step of structural similarity further comprises:
As described below is SSIMiCalculating step, similarly, SSIMjCalculating step be also discussed further below step.
Wherein,Respectively each connected region of blurred picture and clear image
The average brightness value in domain.
Wherein,The respectively brightness scale of each connected region of two images
It is quasi- poor.
Wherein,The brightness related coefficient of connected region is corresponded to for two images.
SSIMi=[li(xi,fi)]a[ci(xi,fi)]b[si(xi,fi)]c, wherein a, b, c are equal.
Wherein, c1, c2, c3It is all larger than zero.Referring specifically to document " Image
quality assessment:from error measurement to structural similarity》(Z.Wang,et
Al.), IEEE Transactios on Image Processing, 2004, vol.13, no.4, pp.600-612.
Wherein, N is the finally obtained connected region number of 1 method of input picture ignorant of the economics deblurring.
Similarly:Wherein, M is the 2 finally obtained connected region of method of input picture ignorant of the economics deblurring
Number.
Input the fuzziness M of blurred pictureblurAre as follows: Mblur=ω1S1+ω2S2, ω1、ω2For different weights.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
In the technical scope disclosed by the present invention, any changes or substitutions that can be easily thought of by those familiar with the art, all answers
It is included within the scope of protection of the present invention.
Claims (6)
1. a kind of video image fuzziness detection method based on Image Blind deblurring, which comprises the following steps:
Step 1: one frame fuzzy video image of input, carries out two distinct types of Image Blind deblurring, respectively obtains clear figure
As f1、f2；
Step 2: by clear image f1、f2Conspicuousness is carried out respectively to detect to obtain notable figure, and notable figure is subjected to connected area segmentation,
And in clear image f1、f2In mark each connected region position；
Step 3: by the fuzzy video image of input and clear image f1、f2Middle connected region same position is marked；
Step 4: calculating separately the fuzzy video image and clear image f of input1、f2The structure of corresponding connected region is similar
Property value SSIMi、SSIMj；Wherein, i is integer and 1≤i≤N, N are Image Blind deblurring side of the input picture through the first seed type
The finally obtained connected region number of method, j is integer and 1≤j≤M, M are that input picture through the Image Blind of second of type removes mould
The finally obtained connected region number of formulating method；
Step 5: calculating the SSIM of all areasiWeighted average S1, SSIMjWeighted average S2,And different weights are assigned as the fuzzy of input blurred picture from S2 to S1
Degree.
2. a kind of video image fuzziness detection method based on Image Blind deblurring according to claim 1, feature
It is, the Image Blind deblurring of the first seed type is specific as follows in the step 1:
(1), blind fuzzy kernel estimates are carried out with image deblurring model and its algorithm based on L1/L2 sparse prior first；
(2), it obtains carrying out deblurring using quickly non-blind deblurring algorithm after blind fuzzy core, obtains clear image f1。
3. a kind of video image fuzziness detection method based on Image Blind deblurring according to claim 1, feature
It is, the Image Blind deblurring of second of type is specific as follows in the step 1:
(1), motion blur algorithm ambiguous estimation core is removed with Fast Blind first；
(2), the non-blind deblurring method of image based on super Laplace prior is recycled, clear image f is obtained2。
4. a kind of video image fuzziness detection method based on Image Blind deblurring according to claim 2, feature
It is, the blind fuzzy kernel estimates further comprise:
(1), derivation is carried out to input fuzzy video image x, obtains high frequency imaging y；
(2), the blind deblurring model modification f constrained based on L1/L2 regular expression sparse prior is utilized1:
K is blind fuzzy core to be estimated, k=[k1,k2,…kv,…,kW], kvFor the component of blind fuzzy core k,
V is the integer more than or equal to 1 and less than or equal to W, and W is the total number of blind fuzzy core to be estimated, and α, β are nonnegative number,
It is f1Derivative,It is1 norm and 2 norms ratio；
F is updated using alternating projection iterative algorithm1, that is, optimize:
(3), weight least square method is assigned again using no constraint iteration and updates blind fuzzy core, that is, optimize:
5. a kind of video image fuzziness detection method based on Image Blind deblurring according to claim 1, feature
It is, the conspicuousness detection method obtains notable figure using GBVS algorithm；The connected area segmentation algorithm uses 8 neighborhoods
Then the binaryzation connected domain figure that segmentation obtains is carried out convolution with obtained clear image and obtains connected region by partitioning algorithm
Figure finally marks out regional location corresponding with the connected region figure in input fuzzy video image.
6. a kind of video image fuzziness detection method based on Image Blind deblurring according to claim 1, feature
It is, the fuzziness M of the input blurred pictureblurAre as follows: Mblur=ω1S1+ω2S2, ω1、ω2For different weights.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610373160.XA CN106097256B (en) | 2016-05-31 | 2016-05-31 | A kind of video image fuzziness detection method based on Image Blind deblurring |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610373160.XA CN106097256B (en) | 2016-05-31 | 2016-05-31 | A kind of video image fuzziness detection method based on Image Blind deblurring |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106097256A CN106097256A (en) | 2016-11-09 |
| CN106097256B true CN106097256B (en) | 2019-01-01 |
Family
ID=57229581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610373160.XA Active CN106097256B (en) | 2016-05-31 | 2016-05-31 | A kind of video image fuzziness detection method based on Image Blind deblurring |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106097256B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106371627A (en) * | 2016-08-26 | 2017-02-01 | 郭曼 | Control system for measuring position information of drawing board |
| CN107067373B (en) * | 2016-12-13 | 2019-08-20 | 南京航空航天大学 | A kind of gradient minimisation recovery method of binary image based on 0 norm |
| CN107767351A (en) * | 2017-09-06 | 2018-03-06 | 郑州工商学院 | A kind of image recovery method of blind deblurring |
| CN109727201A (en) * | 2017-10-30 | 2019-05-07 | 富士通株式会社 | Information processing equipment, image processing method and storage medium |
| CN108154488B (en) * | 2017-12-27 | 2019-10-22 | 西北工业大学 | A kind of image motion ambiguity removal method based on specific image block analysis |
| CN108510453B (en) * | 2018-03-07 | 2021-06-29 | 西安工程大学 | Intelligent traffic monitoring image deblurring method based on visual attention mechanism |
| CN109325939B (en) * | 2018-08-28 | 2021-08-20 | 大连理工大学 | High dynamic image fuzzy detection and verification device |
| CN112927262B (en) * | 2021-03-22 | 2023-06-20 | 瓴盛科技有限公司 | Camera lens shielding detection method and system based on video |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708551A (en) * | 2012-05-17 | 2012-10-03 | 浙江大学 | Image deconvolution method based on super laplace apriori constraint |
| CN103067733A (en) * | 2012-11-28 | 2013-04-24 | 中国科学院研究生院 | Video quality acquisition method |
| CN103413277A (en) * | 2013-08-19 | 2013-11-27 | 南京邮电大学 | Blind camera shake deblurring method based on L0 sparse prior |
| CN103514582A (en) * | 2012-06-27 | 2014-01-15 | 郑州大学 | Visual saliency-based image deblurring method |
| CN104243970A (en) * | 2013-11-14 | 2014-12-24 | 同济大学 | 3D drawn image objective quality evaluation method based on stereoscopic vision attention mechanism and structural similarity |
| CN105184744A (en) * | 2015-08-24 | 2015-12-23 | 西安电子科技大学 | Prior fuzzy kernel estimation method based on standardized sparse measurement image block |
-
2016
- 2016-05-31 CN CN201610373160.XA patent/CN106097256B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708551A (en) * | 2012-05-17 | 2012-10-03 | 浙江大学 | Image deconvolution method based on super laplace apriori constraint |
| CN103514582A (en) * | 2012-06-27 | 2014-01-15 | 郑州大学 | Visual saliency-based image deblurring method |
| CN103067733A (en) * | 2012-11-28 | 2013-04-24 | 中国科学院研究生院 | Video quality acquisition method |
| CN103413277A (en) * | 2013-08-19 | 2013-11-27 | 南京邮电大学 | Blind camera shake deblurring method based on L0 sparse prior |
| CN104243970A (en) * | 2013-11-14 | 2014-12-24 | 同济大学 | 3D drawn image objective quality evaluation method based on stereoscopic vision attention mechanism and structural similarity |
| CN105184744A (en) * | 2015-08-24 | 2015-12-23 | 西安电子科技大学 | Prior fuzzy kernel estimation method based on standardized sparse measurement image block |
Non-Patent Citations (5)
| Title |
|---|
| Blind Deconvolution Using a Normalized Sparsity Measure;Dilip Krishnan 等;《CVPR》;20111231;第233-240页 |
| Image Quality Assessment: From Error Visibility to Structural Similarity;Zhou Wang 等;《IEEE TRANSACTIONS ON IMAGE PROCESSING》;20040430;第13卷(第4期);第600-612页 |
| 基于SSIM与ROI的视频质量评价方法;王正友 等;《仪器仪表学报》;20090930;第30卷(第9期);第1906-1911页 |
| 基于边缘的SSIM图像质量客观评价方法;田浩南 等;《光子学报》;20130131;第42卷(第1期);第110-114页 |
| 视频质量评价方法综述;佟雨兵 等;《计算机辅助设计与图形学学报》;20060531;第18卷(第5期);第735-741页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106097256A (en) | 2016-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106097256B (en) | A kind of video image fuzziness detection method based on Image Blind deblurring | |
| Schick et al. | Improving foreground segmentations with probabilistic superpixel markov random fields | |
| Liang et al. | Counting crowd flow based on feature points | |
| CN110175649B (en) | Rapid multi-scale estimation target tracking method for re-detection | |
| CN109345474A (en) | Image motion based on gradient field and deep learning obscures blind minimizing technology | |
| CN102663436B (en) | Self-adapting characteristic extracting method for optical texture images and synthetic aperture radar (SAR) images | |
| CN109035274B (en) | Document image binarization method based on background estimation and U-shaped convolution neural network | |
| CN105741319B (en) | Improvement visual background extracting method based on blindly more new strategy and foreground model | |
| CN111260738A (en) | Multi-scale target tracking method based on relevant filtering and self-adaptive feature fusion | |
| CN104766065B (en) | Robustness foreground detection method based on various visual angles study | |
| CN103778436B (en) | A kind of pedestrian's attitude detecting method based on image procossing | |
| Liu et al. | Interactive geospatial object extraction in high resolution remote sensing images using shape-based global minimization active contour model | |
| WO2018227882A1 (en) | A priori constraint and outlier suppression based image deblurring method | |
| CN105913392A (en) | Degraded image overall quality improving method in complex environment | |
| CN109146925B (en) | Method for detecting salient target in dynamic scene | |
| Zhou et al. | A robust object tracking algorithm based on SURF | |
| WO2014208963A1 (en) | Apparatus and method for detecting multiple objects by using adaptive block partitioning | |
| CN107067373B (en) | A kind of gradient minimisation recovery method of binary image based on 0 norm | |
| CN106485702A (en) | Image blurring detection method based on natural image characteristic statisticses | |
| CN110751670B (en) | Target tracking method based on fusion | |
| Chen et al. | Visual depth guided image rain streaks removal via sparse coding | |
| CN109767442B (en) | Remote sensing image airplane target detection method based on rotation invariant features | |
| CN108765463B (en) | Moving target detection method combining region extraction and improved textural features | |
| CN107993193B (en) | Tunnel lining image splicing method based on illumination equalization and surf algorithm improvement | |
| CN113421210A (en) | Surface point cloud reconstruction method based on binocular stereo vision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: No. 66, New Model Road, Gulou District, Nanjing City, Jiangsu Province, 210000 Applicant after: NANJING University OF POSTS AND TELECOMMUNICATIONS Address before: 210003 new model road, Nanjing, Nanjing, Jiangsu Applicant before: NANJING University OF POSTS AND TELECOMMUNICATIONS |
|
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161109 Assignee: NANJING NANYOU INSTITUTE OF INFORMATION TECHNOVATION Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2019980001257 Denomination of invention: Video image blur degree detection method based on image blind deblurring Granted publication date: 20190101 License type: Common License Record date: 20191224 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161109 Assignee: Nanjing jinxuetang Information Technology Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2020980006913 Denomination of invention: A method of video image blur detection based on image blind deblurring Granted publication date: 20190101 License type: Common License Record date: 20201020 |
|
| EE01 | Entry into force of recordation of patent licensing contract | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20161109 Assignee: NANJING NANYOU INSTITUTE OF INFORMATION TECHNOVATION Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2021980014141 Denomination of invention: A video image ambiguity detection method based on image blind deblurring Granted publication date: 20190101 License type: Common License Record date: 20211206 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: NANJING NANYOU INSTITUTE OF INFORMATION TECHNOVATION Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2019980001257 Date of cancellation: 20220304 Assignee: Nanjing jinxuetang Information Technology Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2020980006913 Date of cancellation: 20220304 |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: NANJING NANYOU INSTITUTE OF INFORMATION TECHNOVATION Co.,Ltd. Assignor: NANJING University OF POSTS AND TELECOMMUNICATIONS Contract record no.: X2021980014141 Date of cancellation: 20231107 |